SUMMARY Modulators of G-Protein Coupled Receptors (GPCRs) in the human brain have a potential for development of novel treatment strategies targeting neurological disorders such as schizophrenia, Parkinson?s disease, Alzheimer?s disease, and fragile X syndrome/autism. Over the past years more than 10,000 compounds have been identified that interact with muscarinic receptor (mAChRs) and metabotropic glutamate receptor (mGluRs) GPCRs, often allosteri- cally modulating the receptor. Varying pharmacological effects are observed depending on which of several receptor subtypes is engaged and whether the compound is a Positive or Negative Allosteric Modulator (PAM/NAM). The picture is complicated by a number of non-synonymous Single Nucleotide Polymorphisms (nsSNPs) in these recep- tors that are observed in patient populations. It becomes critical to understand when and how a modulator engages the disease mutant receptor as a seemingly minor modification on a scaffold or ?chemotype? may shift selectivity or cause a ?mode switch? between PAM and NAM. However, it is currently not possible to predict how a structural change of the ligand translates into a shift in its pharmacology. It is the central hypothesis of this proposal that a chemotype has an intrinsic ability to bind to a certain allosteric binding pocket in a conserved binding mode and chemical modification on this chemotype dictates selectivity, activity with respect to mutant receptors, or PAM versus NAM activity. With the recently determined experimental structures of both mGluR and mAChR in complex with allosteric modulators we can test this hypothesis. In combination with the breadth and depth of chemical space of known allosteric modulators, it is the objective of this proposal to develop Quantitative Structure-Activity Relation (QSAR) models of allosteric modulation of brain GPCRs. To leverage co-crystal structures as well as small molecule SAR for the construction of such models this proposal develops innovative computational methods that integrate ligand-based (LB) and structure-based (SB) computer aided drug discovery (CADD) methods. I will map QSAR models onto structural models of the allosteric modulator in complex with the GPCR and so highlight the structural determinants of activity. Selected ligands will be co-crystallized with the receptor to critically evaluate and ultimately confirm the computational modeling approaches and facilitate CADD. In collaboration, I will demonstrate that such models spur the development of lead and probe compounds with tailored pharmacological profiles that help study the biological function of these receptors. Compu- tational models will be confirmed in an iterative feedback loop through mutagenesis studies and co-crystallization through collaboration partners. Ultimately, they will become starting points for a second generation of allosteric mod- ulators with the mode of action that is understood at atomic level of detail.